Microsoft models
Microsoft has 12 models in the directory. The newest is MAI-Image-2.6, released Oct 13, 2026.
The cheapest to run is Phi 4 Multimodal at $0.07 per million input tokens.
Phi 4 Multimodal has the longest context window, 128K tokens.
- Models
- 12
- Newest
- Oct 13, 2026
- From
- $0.07/M input
- Longest context
- 128K
All Microsoft models
Filter in the directoryMAI-Image-2.6
ImageGenerates images from text and images.
by microsoftOct 13, 2026
MAI-Image-2.6-Flash
ImageGenerates images from text and images.
by microsoftOct 13, 2026
MAI-Transcribe-2-Streaming
TranscriptionTurns speech into text.
by microsoftOct 1, 2026
MAI-Voice-2.1
SpeechTurns text into spoken audio.
by microsoftOct 1, 2026
MAI-Voice-2.1-Flash
SpeechTurns text into spoken audio.
by microsoftOct 1, 2026
MAI-Transcribe 2
TranscriptionTurns speech into text.
by microsoftSep 3, 2026
MAI-Voice-2-Flash
SpeechTurns text into spoken audio.
by microsoftJul 23, 2026
MAI-Voice-2
SpeechTurns text into spoken audio.
by microsoftJun 2, 2026
- 128K context
Reads text.
by microsoftFeb 26, 2025$0.07/M input tokens$0.11/M output tokens
- 16K context
Reads text. Structured output. Served by 3 providers.
by microsoftDec 11, 2024$0.125/M input tokens$0.50/M output tokens3 providers
- 8K context
Reads text. Tool calling. Served by 3 providers.
by microsoftDec 11, 2024$0.075/M input tokens$0.30/M output tokens3 providers
- 66K context
Reads text. Structured output. Served by 3 providers.
by microsoftApr 15, 2024$0.62/M input tokens$0.62/M output tokens3 providers
Prices and context from models.dev (MIT licensed), refreshed Nov 18, 2026.